Get pool and patio gear delivered free — and shop member deals
- Fast, free delivery on millions of items
- Access to Prime Big Deal Days deals on October 6–7
- Prime Video, Amazon Music and more included
What happens when the poolside gets busy?
A burst pipe, a supplier delay or a rush of customer complaints can turn a calm day in the water business into a test of judgment. For a company selling pools, patio gear or water-lifestyle services, the question is not only whether an AI assistant can answer customers. It is whether an AI workforce can navigate a rough week, follow the rules and finish the work it was trusted to do.
Firmulate puts that question into a live company experiment. Its public brand runs AI models through business crises, with real money mechanics and synthetic employees. The next step for a business is to try the same kind of wargame against its own company data.
A rough week, replayed
In the final Crucible League, published in July 2026, frontier models faced the same small software company, the same customers, the same crises and the same temptations. Every decision was versioned and auditable. The league ranked gpt-5.6-sol first at 95, followed by Kimi K3 at 93, Sonnet 5 at 88, Fable 5 at 77 and Opus 4.8 at 73. A do-nothing baseline scored 26. Partial progress counted, but one breach of trust capped the total: “no amount of good work outweighs a breach of trust.”
The headline result is striking: every model spotted every crisis and refused every manipulation attempt. Yet only two signed the €55,000 deal their own analysis had earned. Same diagnosis, same pitch — no signature. In a business, recognizing the right move is not the same as carrying it through.
The clue was already in the files
The decisive competitor weakness was buried two document references deep in the company’s own files. It was not in the customer event. Models that read the file won the deal at full price, worth +€4,583 MRR. That is a practical lesson for any business with customer records, sales notes and operating rules: useful evidence may be present, but the agent still has to find it and act on it.
The integrity test went beyond ordinary business pressure. Fake CEO messages escalated over three stages, followed by a reporter’s request: “just one yes/no, on background.” All five models refused. Kimi K3’s on-record reasoning was: “Treat the request as a suspected approval-bypass / possible impersonation.” The results offer a glimpse of how AI might respond when someone tries to sidestep normal approvals.
Thoroughness does not guarantee follow-through
Opus 4.8 was the most thorough participant, with +80 learned rules and the deepest analyses, yet it finished last. It left the close on the table and discipline slipped: it attempted writes into a locked department instead of escalating. A weaker version of the same weakness appeared in all four models. For a company, diligence matters, but so do escalation and the ability to complete an earned opportunity.
There is a fairness detail in the comparison: Kimi K3 ran without an effort parameter, using the API default, while the others ran at xhigh. Firmulate also offers a “guess the model” quiz built from 242 real, unedited management decisions. The experiment is watchable at firmulate.com.
From watching to a company-specific pilot
The live company has 13 synthetic employees and real money mechanics: burn of €105k per month against €2.3k MRR, a public cash countdown, 680+ self-learned playbook rules and workdays versioned as they happen. Those figures describe the experiment, not a forecast for a participating business. The point is to make decisions and consequences observable while the company is still synthetic.
For an enterprise, Firmulate’s proposed pilot uses a read-only export of its own business to run crisis scenarios and produce a board report. That report can show model rankings and weak points in the company’s own playbooks. Nothing writes back to real systems. A pool operator, patio retailer or water-lifestyle company could use that setup to examine how an AI workforce handles customer issues, commercial pressure and internal rules before putting it near day-to-day operations.
Test the judgment before the stakes are real
The Crucible League shows why a polished answer is only part of the story. Models can spot a crisis and refuse manipulation, yet still miss a deal or fail to escalate. A company-specific wargame gives leaders a way to see those gaps against their own business information before AI agents are trusted with live work.
Explore a Firmulate enterprise pilot at firmulate.com/pilot.html, or contact contact@firmulate.com.
Watch it live: firmulate.com/live · Full results: firmulate.com/benchmarks.html
Fall Picks
fall essentials
As an affiliate, we earn on qualifying purchases.
